Compare AI models
| Specifications | ||
ParametersThe learned values that shape how a model responds. B means billion and T means trillion. For mixture-of-experts models, the total count is larger than the number active for each token. Parameter count alone does not measure quality. | 109B | — |
Context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider. | 10M | — |
| API pricingUSD per 1M tokens · lower wins | ||
Cheapest inputLowest input rate across every provider on OpenRouter, the lab included. The cheapest endpoint may run a quantised build or a shorter context — see "Available from" on the model page. | $0.08DeepInfra | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $0.24DeepInfra | — |
| $35.00DeepInfra · fp8 | — | |
| Benchmarks | ||
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | 74.3% | 72% |
MathVistaMultimodal reasoning — mathematical reasoning about visual information | 70.7% | 68.1% |
MMMUMultimodal — Tests the AI on understanding images and text together across many college subjects. Higher is better. | 69.4% | 63.2% |
BullshitBench v2Nonsense detection — Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better. | 20% | — |
LiveCodeBenchCompetitive coding — Coding problems published so recently the AI can't have seen them in training — a contamination-free test of raw programming skill. Higher is better. | 32.8% | — |
HumanEvalFunction synthesis — 164 small Python problems: the AI is given a function's description and has to write the working function. This was the coding benchmark of the GPT-3.5 and GPT-4 era, before the field moved to fixing real bugs in real repositories. Higher is better. | — | 85.7% |
MATHReasoning — accuracy on the original competition mathematics test set | — | 73% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 57.2% | — |
MMLUGeneral knowledge — A 57-subject multiple-choice exam — history, law, medicine, maths — that was the standard measure of how much a model knows from 2020 until roughly 2024, when frontier scores crowded into the high 80s and labs moved on to harder tests. The scores here were published years apart under different testing setups, so read them as a historical record rather than a like-for-like ranking. Higher is better. | — | 86.2% |
MGSMMultilingual math — solving grade-school word problems translated into multiple languages | 90.6% | — |
MMMU-Pro · standard and vision meanMultimodal reasoning — average across the standard and vision sets | 52.2% | — |
| Overview | ||
| Company | Meta | SpaceXSI |
| Release date | Apr 5 2025 | Aug 13 2024 |
| Access | Open Weight | Closed |
| Model details | View model | View model |
Other comparisons
LLaMA 4 ScoutvsClaude Fable 5.2Grok‑2 minivsClaude Fable 5.2LLaMA 4 ScoutvsGPT-6.1 TerraGrok‑2 minivsGPT-6.1 TerraLLaMA 4 ScoutvsGemini 4 HydrogenGrok‑2 minivsGemini 4 HydrogenLLaMA 4 ScoutvsDeepSeek-V4.1-Flash-1017Grok‑2 minivsDeepSeek-V4.1-Flash-1017LLaMA 4 ScoutvsMistral Large 4Grok‑2 minivsMistral Large 4LLaMA 4 ScoutvsKimi K3.1 ProGrok‑2 minivsKimi K3.1 ProFrequently asked questions
LLaMA 4 Scout leads Grok‑2 mini on 3 of the 3 benchmarks they both report (MMLU-Pro, MathVista, MMMU). Grok‑2 mini shipped 235 days before LLaMA 4 Scout, so benchmark comparisons should account for the intervening progress.
LLaMA 4 Scout is open weight, while Grok‑2 mini is closed.
On MMLU-Pro, LLaMA 4 Scout leads at 74.3% vs Grok‑2 mini at 72%. On MathVista, LLaMA 4 Scout leads at 70.7% vs Grok‑2 mini at 68.1%. On MMMU, LLaMA 4 Scout leads at 69.4% vs Grok‑2 mini at 63.2%.